Cross, H. A., D. Cavanaugh, C. C. Buonanno, and A. Hyman, 2021: The Impact of the Storm Prediction Center’s Convective Outlooks and Watches on Emergency Management Operational Planning. J. Operational Meteor., 9 (3), 36-46, doi: https://doi.org/10.15191/nwajom.2021.0903.
Short Contribution
The Impact of the Storm Prediction Center’s Convective Outlooks and Watches on Emergency Management Operational Planning
HEATHER A. CROSS, DENNIS CAVANAUGH, CHRISTOPHER C. BUONANNO NOAA/National Weather Service, Little Rock, AR
AMY HYMAN Arkansas State University, Jonesboro, AR
(Manuscript received 11 October 2020; review completed 16 March 2021)
ABSTRACT For many emergency managers (EMs) and National Weather Service (NWS) forecasters, Convective Outlooks issued by the Storm Prediction Center (SPC) influence the preparation for near-term severe weather events. However, research into how and when EMs utilize that information, and how it influences their emergency operations plan, is limited. Therefore, to better understand how SPC Convective Outlooks are used for severe weather planning, a survey was conducted of NWS core partners in the emergency management sector. The results show EMs prefer to wait until an Enhanced Risk for severe thunderstorms is issued to prepare for severe weather. In addition, the Day 2 Convective Outlook serves as the threshold for higher, value-based decision making. The survey was also used to analyze how the issuance of different risk levels in SPC Convective Outlooks impact emergency management preparedness compared to preparations conducted when a Convective Watch is issued.
1. Introduction The Convective Outlooks have continued to change and evolve through the years. For example, Convective Outlooks are routinely issued several in 2012, an increase to the temporal resolution of the times a day by the Storm Prediction Center (SPC). These Convective Outlooks was researched and implemented outlooks highlight areas across the country where both experimentally without adding a larger workload for severe and non-severe thunderstorms are anticipated. forecasters (Jirak et al. 2012). Regardless of changes to Areas of possible risk from severe thunderstorms are the product, SPC Convective Outlooks have grown in categorized as “marginal risk”, “slight risk”, “enhanced popularity since their origins in the early 1950s (Corfidi risk”, “moderate risk”, or “high risk”, depending on the 1999). With a growing interest in the SPC Convective coverage and strength of the expected severe storms. Outlooks, emergency managers (EMs) have started “High risk” days, while rarely used, tend to verify well. to reference the information as it pertains to their A study of “high risk” forecasts from 2003 to 2009 jurisdiction in the days and hours leading up to severe showed around 64% of those forecasts verified on weather (Ernst et al. 2018). Additionally, it has been tornado warnings alone (Davis et al. 2010). Before the shown that EMs consult the Convective Outlooks early outlooks are published, forecasters at the SPC perform in the planning phase before severe weather occurs a detailed analysis of recent and current weather data (Baumgart et al. 2008). While the National Weather and forecast model output to produce an outlook Service (NWS) has the task of sending hazardous representative of the current and unfolding atmosphere weather information out to the public, it is the role of (Storm Prediction Center 2020). EMs to take the necessary precautions to keep those in
Corresponding author address: Dennis Cavanaugh, 8400 Remount Rd, North Little Rock, AR 72118. E-mail: [email protected] 36 Cross et al NWA Journal of Operational Meteorology 11 June 2021 their jurisdiction safe. The EMs take the necessary steps contact with the NWS and EMs (Hoss and Fischbeck to do this by planning, preparing, and practicing when 2018). This suggests that, while it is very easy to find the weather is quiet. This allows them to implement and view the SPC Convective Outlook with today’s mitigation techniques when hazardous weather is likely, technology, understanding outlooks might not be as or imminent, and to build community resilience to the simple, especially for those EMs unfamiliar with – or impacts of hazardous weather (Morss and Ralph 2007). without an educational background in – meteorology When severe weather approaches, EMs respond in a (Weaver et al. 2014). Furthermore, even with a firm variety of ways from activating notification systems and understanding of the Convective Outlooks, how an Emergency Operation Centers (EOCs), to forwarding EM uses those in the planning and preparation phase critical weather warning and additional key information leading up to a severe weather event is unknown. to the public (League et al. 2010). This research aims to determine when in Days 1 Convective Watches, classified as either a Severe to 8 leading up to a severe weather event, and at what Thunderstorm Watch (SVA) or a Tornado Watch Convective Outlook level, EMs begin altering their (TOA), are issued by the SPC when organized severe normal routine to prepare for hazardous weather. This weather is expected within a two to eight hour period. study also explores how EMs alter their operations Convective Watches were first issued in 1952 when the when a Convective Watch is issued, and whether or not Weather Bureau Army Navy (WBAN) Analysis Center the type of watch impacts those changes. was established. The WBAN Severe Weather Unit, that through the years would eventually evolve into the 2. Data and methods current day SPC, was responsible for the issuance of weather bulletins, which are essentially what we know A survey was developed to better understand when, today as Convective Watches (Corfidi 1999). An SVA is and at what Convective Outlook level, EMs enact issued when thunderstorms capable of producing hail of their severe weather operations plan to change their 1-in diameter or larger and/or damaging thunderstorm routine and prepare for hazardous weather based on winds are expected, and a TOA is issued when large hail information from the SPC. The SPC is also responsible and damaging winds are possible, as well as tornadoes for issuing Convective Watches to alert the public and (Storm Prediction Center 2021). Once in a while, the EMs when conditions are favorable for the development SPC will include the “Particularly Dangerous Situation of severe thunderstorms and/or tornadoes within the (PDS)” statement in TOAs when long-lived and intense proceeding hours (Storm Prediction Center 2020), so tornadoes are likely. Additionally, the PDS Watches are pertinent questions were included in the survey. This typically issued when there is a “high risk” included grouping also included PDS Watches to determine what in the Day 1 Convective Outlook (Storm Prediction difference in planning, if any, this would cause when Center 2021). issued compared to a typical Convective Watch. In addition to the protection of life, the NWS and The survey was created and disseminated online agency core partners have the mission of mitigating in the spring of 2018 using Google Forms, and can be the loss of property attributable to hazardous weather viewed in Appendix A. The survey link was sent via events. Historic events show that the dissemination email to EMs and four other NWS Warning Coordination of critical weather information leading up to severe Meteorologists, who forwarded the survey link to EMs in weather can save millions of dollars (Grice et al. 1999). their forecast area. In all, the survey was disseminated to Previous research by League et al. in 2010 shows that approximately 160 state, county, city, and private sector the majority of EMs use NWS products, especially EMs throughout Arkansas, Oklahoma, and portions watches and warnings. However, research into how of Texas and Tennessee. Eighty-four EMs completed EMs prepare in the Day 1 to 8 period leading up to and returned the survey, resulting in a response rate a severe weather event, and how products issued by just ≥50%. While the survey was primarily distributed the SPC impact that, is lacking. Through the years, online, it was also handed out at an Integrated Warning SPC’s Convective Outlooks have transitioned from a Team meeting to Arkansas EMs. Around 10 surveys generic tornado forecast into the risk categories ranging were completed on paper, and then entered into the from “general thunder” to “high” that we know today Google Form manually by the authors in order to (Corfidi 1999). Past studies show that the best method compile all responses. Google Forms allows users to for managing weather information is through direct adjust settings within the survey before sending it out.
ISSN 2325-6184, Vol. 9, No. 3 37 Cross et al NWA Journal of Operational Meteorology 11 June 2021
For this research, email addresses were not collected, Watches included watching the weather more closely/ users were not required to sign in, and there was no monitoring radar, sending out information, beginning limit on responses. Additionally, respondents could not briefings, cancelling events, watching local news for edit their answers after submission, nor could they view weather information, calling the local NWS/monitoring summary charts and text responses. NWSChat, deploying trained weather spotters, and no To simplify the process of data acquisition and impact. analysis, the team decided to use multiple-choice As mentioned above, one of the goals of the questions rather than ones that would lead to open- survey was to determine what risk level on Days 1 to ended responses. Although allowing EMs to answer 8 prompts EMs to take action. However, the team also with an open-ended response would provide more wanted to analyze the value of the decisions made by information overall, it was decided that closed-ended the EM, because in most cases this plays a critical role survey questions would meet the researchers’ goal of in determining what action is taken. In particular, when identifying thresholds and significant differences in the financial implications are in play it can be especially decisions EMs make using the SPC’s products, and thus stressful for decision makers because they could be open were ultimately used instead of open-ended questions. to steeper criticism depending on how the event plays In proceeding with the multiple-choice questions, out (Ernst et al. 2018). To assess how much attention an authors included ways for EMs to opt out of answering EM gives to a weather event, (1s) and (2s) were assigned questions. While “other” was not an answer choice for to the answer choices for each question. A (1) indicated any survey questions, respondents could choose not to that the decisions made required no additional financing answer questions that were not required, and for those or resources, whereas a (2) indicated that decisions made that were required, a final option of “I don’t use SPC required additional money and resources. A zero was Convective Outlooks” or “No impact” were included. assigned to those responses that indicated “no impact”. Moreover, survey respondents were anonymous, so Because more than one decision could be made because they had nothing to lose or gain by selecting an option of the “check all that apply” format, the sum of each that did not correctly reflect their operations in the given value-weighted decision chosen was taken and used for scenarios. Finally, in an attempt to limit potential NWS statistical analysis for each applicable question. perception and bias in the survey that could influence Although the method above allowed researchers to the results, a professor of emergency management from gauge how each outlook impacted the EM’s focus on a Arkansas State University was consulted to create and weather event, it didn’t necessarily accurately reflect the finalize the survey. financial weighted decisions. This is because multiple The survey included questions such as state answers could be chosen. Therefore, it was possible for residency, emergency management jurisdiction/size, an EM to choose three answers rated a (1) with a sum of and primary sources for weather information. However, 3, while another EM could have chosen only one option the bulk of the survey consisted of specific questions rated a (2), making it seem that more resources were pertaining to emergency operations plans for severe spent with the sum of 3 rather than the sum of 2. To weather based on the SPC Convective Outlooks and resolve this problem, researchers re-ran statistical tests Watches. These questions were intended to discover and changed the (1s) to (0s) and (2s) to (1s). This would if and how each SPC product affects operations result in a sum of (0) unless the choices that required (e.g., differences in planning, if any, caused by an additional money and resources were chosen (1s). “enhanced risk” as opposed to a “moderate risk” on Day 1; differences in resource allocation caused by a 3. Discussion and analysis SVA compared to a TOA, etc.). When asked how their operations would be impacted by Convective Outlooks, In all, there were 84 responses to the survey, 29 from EMs could choose watching the weather more closely, Arkansas, 18 from Tennessee, three from Texas, and 34 sending out additional information, beginning daily from Oklahoma. The majority of the responses were briefings, cancelling events, and changing staffing from county EMs, however there were several state levels. However, these questions were not required and city EMs represented, and two private enterprise and EMs were only asked to answer these particular EMs represented as well. Jurisdiction population sizes questions if they used SPC Convective Outlooks. ranged from ≤2500 to ≥3 000 000. Although most EMs Additionally, response choices regarding Convective who took the survey indicated a jurisdiction size ≤2500,
ISSN 2325-6184, Vol. 9, No. 3 38 Cross et al NWA Journal of Operational Meteorology 11 June 2021 almost a quarter reported a population size between use SPC Outlooks wait to make high value decisions 300 000 and 999 999. In addition to other generalized until their area is at least under a “slight risk” in the questions, it was of interest to determine how many EMs Days 1 to 3 period, but that approximately half prefer are even aware of the SPC’s Convective Outlooks. As to wait until their area is under an “enhanced risk” or seen in Fig. 1, of those surveyed, 81% answered “Yes” greater. to this question, leaving 19% who answered “No”. EMs As discussed above, the authors assigned (2s) to were also asked if they had an emergency operation any decisions EMs made in Days 1 to 8 leading up to plan for severe weather, which 98% of those surveyed a severe weather event that required additional money did. Survey results, seen in Fig. 2, show that for 66% or resources. The sum of these were then incorporated of EMs, their main source of weather information into a Kruskal-Wallis test to analyze which day had comes from the local NWS Weather Forecast Office, the greatest impact on EM decision making (Table 1). with weather apps on a smartphone the second most The results support the finding that EMs start making common source of weather information. The SPC, was most of their higher value-based decisions on Day 2. well represented, but only the main source of weather Therefore, the Convective Outlook issued by the SPC information for roughly 12% of EMs, meaning that on Day 2 plays a critical role in how EMs prepare for a although they view SPC’s products, it isn’t necessarily severe weather event. where they go first or most frequently for weather information. It is important to mention, however, that many NWS offices utilize, agree with, and message the SPC Convective Outlooks in their products. Therefore, even if EMs do not go to SPC directly as a primary source, they are still using the SPC information via their local forecast office. To analyze the survey results with respect to those EMs that use SPC Outlooks, a Kruskal-Wallis test [Weiss (2008); Albright and Winston (2015); Stat Tools (2020)] was performed to determine what level of SPC Outlook prompted/impacted decision making. A Kruskal-Wallis test is a non-parametric method used for comparing two or more independent samples of equal or different sample sizes. We wanted to test three independent samples of SPC Outlooks (e.g., Days 1 to Figure 1. Bar graph showing the spread of how many 3) based on EM responses, therefore, the Kruskal-Wallis EMs are aware of Storm Prediction Center Convective test was chosen because the distribution of results was Outlooks. Click image for an external version; this not likely to fall into a normal (Gaussian) distribution. applies to all figures hereafter. This non-parametric test is a variation of a one-way analysis of variance (ANOVA), but is not based on the assumption that the sample size must be normally distributed to derive statistically significant results. Of the 84 EMs surveyed, 16 indicated that they were not aware of SPC Outlooks, and thus did not use them to make any planning decisions. This left 68 respondents to use for the Kruskal-Wallis test to analyze at what categorical risk level EMs started making some value- based decisions (Fig. 3). This test showed that EMs who use SPC Outlooks to make value-based decisions on average made those decisions when their jurisdiction was in an “enhanced risk” area (4 out of 6) based on SPC’s threat categories. Furthermore, results from the survey in Fig. 4 show that roughly 90% of the EMs who Figure 2. Bar graph showing the main source of weather information used by EMs.
ISSN 2325-6184, Vol. 9, No. 3 39 Cross et al NWA Journal of Operational Meteorology 11 June 2021
Table 1. Kruskal-Wallis Test analyzing the threshold at which EMs stated in the survey they started making decisions on Days 1 to 8. The results show that EMs begin to make value-based decisions on Day 2, which places high importance on that convective outlook. Day 1 Day 2 Day 3 Days 4 to 8 Decision Decision Decision Decision Threshold Threshold Threshold Threshold Sample Size 68 68 68 68 Sample Mean 0.44 0.38 0.00 0.00 Sample Standard Deviation 0.90 0.86 0.00 0.00 Sample Median 0 0 0 0
In addition to the Convective Outlook period, EMs were also asked about decisions they make during the Convective Watch phase of an anticipated thunderstorm event. EMs were asked to list their decisions made for the following types of Convective Watches; a SVA, a TOA, and a TOA with a PDS designation. Because watches are not ranked by severity (e.g., the probability metadata included within the watch issuance on the SPC watch page), a non-parametric hypothesis test was performed for the pairs of watch types to determine if there was a statistically significant difference in decision making by the EM community based on the type of watch issued. Mann-Whitney tests [Mann and Figure 3. Box and whisker plot representing the results Whitney, 1947, Weiss, 2008, and Stat Tools, 2020] were of a Kruskal-Wallis test with Day 1 depicted in orange, performed for each pairing of watch type, and for each Day 2 in blue, and Day 3 in green. The test shows that pair, these tests indicated that there was a statistically Day 1 compared to Day 3 reflects the largest difference significant difference in the amount of decisions made between high value decision-making thresholds. based on the type of watch that is issued preceding However, it also shows that most EMs start making a high-impact thunderstorm event (Tables 2-4). A decisions to change their operations or start monitoring Kruskal-Wallis test was also conducted using the the weather more closely when the SPC outlook reaches method with (0s) and (1s) to take the findings further a level 4 out of 6 (i.e., an “enhanced risk”). and determine if EMs are also spending more resources for a TOA and PDS Watch than an SVA. The Kruskal- Wallis test showed that, in fact, this was true (Table 5). These results indicate that EMs change their operations more and make higher value-based decisions based on the type of Convective Watch issued. Moreover, the Mann-Whitney test results provide strong evidence that EMs pay more attention and spend more resources when a TOA is issued compared to an SVA. Finally, results show that EMs are more focused on PDS Watches when compared to TOAs or SVAs. This suggests the Figure 4. Clustered column chart showing at which importance of forecasting and communicating the outlook level on Days 1 to 3 emergency operations are convective threat severity in the hours leading up to a impacted. hazardous weather event. Last, EMs responded to the survey from a large disparity of populations served. In general, EMs responsible for a larger population area tend to have
ISSN 2325-6184, Vol. 9, No. 3 40 Cross et al NWA Journal of Operational Meteorology 11 June 2021
Table 2. Mann-Whitney test results using the (1) and (2) method comparing the decision-making for SVAs compared to TOAs. SVA Decision Value TOA Decision Value Sample Size 84 84 Sample Mean 2.93 4.91 Sample Standard Deviation 2.24 2.72 Sample Median 3 5 Mann-Whitney Test Null Hypothesis Neither Dist. Smaller Alternative Hypothesis SVA Decision Value Smaller Number of Values in Ranking 168 Number of Tied Values 168 Rank Sum for Variable 1 5588.5 Rank Sum for Variable 2 8607.5 Normal Approximation (NA) Used Yes Ties Present, but Not Corrected For No Mean for NA 7098 Standard Deviation for NA with Tie Correction 312.74 z-Statistic for NA with Tie Correction –4.83 p-Value <0.0001 Null Hypothesis at 10% Significance Reject Null Hypothesis at 5% Significance Reject Null Hypothesis at 1% Significance Reject more resources available to deploy if a high-impact and associated p-value indicate that population size is weather event affects the community they serve. In not a good indicator of value-based decision-making lieu of asking EMs how much money they spent in by EMs. This regression provides strong support that terms of time, personnel, and resources, the survey population size and resource availability did not skew asked for relative decisions being made that may result the results of the relative value of decision making by in a response to an impactful weather event. This EMs that responded to this survey. method was used to avoid quantifying a large financial disparity from EMs representing smaller communities 4. Summary and conclusion compared to larger communities. The relative cost of dedicating resources is assumed to be proportional to The EMs that responded to this survey are using community size, resources, and money available. To the SPC Convective Outlook and Convective Watch test this hypothesis, the relative value-based decision- products to help prepare for hazardous weather, and making tabulated from the survey was compared to this research provides novel insight into how the EM EMs representing a multitude of population categories community uses those products to prepare communities using a linear regression. With this technique, the sum before and during severe weather events. One limitation of each value-based decision for a PDS Convective of this research included a smaller sample size than Watch was represented as the dependent variable originally desired, with only 84 respondents. However, leaving population size served as the independent completed surveys still provided an adequate amount variable (Table 6). The PDS was the best option for the of data for analysis and thus resulted in worthwhile regression test (compared to Outlook Day 1, TOA, SVA, and meaningful statistical results. Although EMs etc.) because it resulted in the highest amount of value- from several different states were included in the based decisions. The resultant regression coefficient survey dissemination, a larger sample size from Texas,
ISSN 2325-6184, Vol. 9, No. 3 41 Cross et al NWA Journal of Operational Meteorology 11 June 2021
Table 3. Mann-Whitney test results using the (1) and (2) method comparing the decision-making for TOAs compared to PDS TOAs. SVA Decision Value TOA Decision Value Sample Size 84 84 Sample Mean 4.91 6.18 Sample Standard Deviation 2.72 3.11 Sample Median 5 6 Mann-Whitney Test Null Hypothesis Neither Dist. Smaller Alternative Hypothesis TOA Decision Value Smaller Number of Values in Ranking 168 Number of Tied Values 167 Rank Sum for Variable 1 6261 Rank Sum for Variable 2 7935 Normal Approximation (NA) Used Yes Ties Present, but Not Corrected For No Mean for NA 7098 Standard Deviation for NA with Tie Correction 313.54 z-Statistic for NA with Tie Correction –2.6679 p-Value 0.0038 Null Hypothesis at 10% Significance Reject Null Hypothesis at 5% Significance Reject Null Hypothesis at 1% Significance Reject
Oklahoma, and Tennessee could have allowed the data versus a TOA, as well as the added significance of the to be broken down even further. If this is attained in PDS designation on EM preparedness. If the NWS and future research, the authors would aim to determine EM communities are aware of these thresholds in value- what impact, if any, location plays on EM operational based decision-making, they may be able to optimize planning in response to the SPC products. resources while maintaining a high level of community When the NWS better understands the decision resilience to hazardous weather events. thresholds of their core partners, they can provide forecasts and briefings that target important emergency Acknowledgments. Thank you to NWS Norman, operations plans and this would help all partners to NWS Fort Worth/Dallas, NWS Memphis, and NWS collaborate on improving community resilience. This Nashville for disseminating the survey to EMs in their is especially helpful for EMs and the NWS regarding forecast areas, which helped widen the sample size and threats in the Convective Outlooks. For example, when improve research analysis. The authors also thank all local NWS offices know EMs prefer to wait until an the EMs who took the time to take the survey, and the Enhanced Risk is issued in the Convective Outlook to reviewers who helped improve the manuscript through prepare for severe weather, it can directly impact when thoughtful comments and suggestions. key information is relayed or presented during briefings. This research also helps the NWS and EM communities —————————— identify thresholds in which finite resources are being used. This is primarily shown by the importance of the Day 2 Convective Outlook as a common threshold for higher value-based decision-making by EMs. It is also reflected in the SVA impact on operational planning
ISSN 2325-6184, Vol. 9, No. 3 42 Cross et al NWA Journal of Operational Meteorology 11 June 2021
Table 4. Mann-Whitney test results using the (1) and (2) method comparing the decision-making between SVAs and PDS TOAs. SVA Decision Value TOA Decision Value Sample Size 84 84 Sample Mean 2.93 6.18 Sample Standard Deviation 2.24 3.11 Sample Median 3 6 Mann-Whitney Test Null Hypothesis Neither Dist. Smaller Alternative Hypothesis SVA Decision Value Smaller Number of Values in Ranking 168 Number of Tied Values 168 Rank Sum for Variable 1 5006 Rank Sum for Variable 2 9190 Normal Approximation (NA) Used Yes Ties Present, but Not Corrected For No Mean for NA 7098 Standard Deviation for NA with Tie Correction 313.41 z-Statistic for NA with Tie Correction –6.67 p-Value <0.0001 Null Hypothesis at 10% Significance Reject Null Hypothesis at 5% Significance Reject Null Hypothesis at 1% Significance Reject
Table 5. Kruskal-Wallis Test using the (0) and (1) method showing that EMs make higher value-based decisions when a TOA or PDS Watch is than they do when an SVA is issued. SVA Decision TOA Decision PDS Decision Threshold Threshold Threshold Sample Size 84 84 84 Sample Mean 0.19 0.80 1.30 Sample Standard Deviation 0.40 0.88 1.06 Sample Median 0 1 1
ISSN 2325-6184, Vol. 9, No. 3 43 Cross et al NWA Journal of Operational Meteorology 11 June 2021
Table 6. Regression analysis with all emergency management value-based decisions set as dependent variables for PDS SPC Watches. The independent variable is population size represented by the EMs. The p-Value associated with each independent variable suggests that population size represented was not a good discriminator of relative value-based decisions. These results suggest that relative value-based decisions were not dependent upon population size represented. Regression for PDS Watch Standard Adjusted Error of Rows Ig- Rows Ig- Multiple R R-Square R-Square Estimate nored nored Summary 0.07 0.01 0.00 2.13 0 0 ANOVA Table Degrees of Sum of Mean of Freedom Squares Squares F p-Value Explained 1 1.77 1.77 0.39 0.53 Unexplained 82 372.47 4.54 Regression Table Decision Standard Confidence Interval 95% Value Coefficient Error t-Value p-Value Lower Upper Constant 2.91 0.56 5.19 <0.0001 1.80 4.03 Population –0.12 0.19 –0.62 0.53 –0.50 0.26
APPENDIX A
Survey Questions
Impacts of the Storm Prediction Center’s Convective Outlooks on Emergency Management (* Required)
1. What state are you in? *
2. What are you an Emergency Manager of? * □ Medical Facility □ State □ Event Center □ City □ County □ Educational Facility □ Shopping Center
3. On any given day, what population size do you serve? * □ <2500 □ 2,500 - 49,999 □ 50,000 - 99,999 □ 100,000 - 999,999 □ 1,000,000 - 3,000,000 □ >3 Million
4. What is your primary source for weather information? * □ Weather apps on a smart phone □ Television □ Storm Prediction Center □ Social Media □ Radio □ Local National Weather Service
5. Do you have an Emergency Operations Plan for severe weather? * □ Yes □ No
ISSN 2325-6184, Vol. 9, No. 3 44 Cross et al NWA Journal of Operational Meteorology 11 June 2021
6. Are you aware of the Storm Prediction Center’s (SPC) Convective Outlooks? * □ Yes □ No
7. If on Day 1, the area that you serve is highlighted by SPC in a Convective Outlook, at which outlook level would your operations procedure be impacted? * □ General Thunder □ Marginal □ Slight □ Enhanced □ Moderate □ High □ I don’t use SPC Convective Outlooks
7a. How? (Please answer if you use SPC Convective Outlooks) □ Watch the weather forecast more closely □ Send out information □ Begin daily briefings □ Cancel events □ Change staffing levels
8. If on Day 2, the area that you serve is highlighted by SPC in a Convective Outlook, at which outlook level would your operations procedure be impacted? * □ General Thunder □ Marginal □ Slight □ Enhanced □ Moderate □ High □ I don’t use SPC Convective Outlooks
8a. How? (Please answer if you use SPC Convective Outlooks) □ Watch the weather forecast more closely □ Send out information □ Begin daily briefings □ Cancel events □ Change staffing levels
9. If on Day 3, the area that you serve is highlighted by SPC in a Convective Outlook, at which outlook level would your operations procedure be impacted? * □ General Thunder □ Marginal □ Slight □ Enhanced □ Moderate □ High □ I don’t use SPC Convective Outlooks
9a. How? (Please answer if you use SPC Convective Outlooks) □ Watch the weather forecast more closely □ Send out information □ Begin daily briefings □ Cancel events □ Change staffing levels
10. If on Days 4-8, the area you serve is highlighted by SPC in a Convective Outlook, at which outlook level would your operations procedure be impacted? * □ 15% □ 30% □ I don’t use SPC Convective Outlooks
10a. How? (Please answer if you use SPC Convective Outlooks) □ Watch the weather forecast more closely □ Send out information □ Begin daily briefings □ Cancel events □ Change staffing levels
11. How are your operations impacted when the area you serve is placed under a Severe Thunderstorm Watch? * □ Watch the weather forecast more closely/Monitor Radar □ Send information out □ Begin briefings □ Cancel events □ Change staffing levels □ Watch local news for weather information □ Call your local National Weather Service/Monitor NWSChat □ Deploy trained weather spotters □ No impact
ISSN 2325-6184, Vol. 9, No. 3 45 Cross et al NWA Journal of Operational Meteorology 11 June 2021
12. How are your operations impacted when the area you serve is placed under a Tornado Watch? * □ Watch the weather forecast more closely/Monitor Radar □ Send information out □ Begin briefings □ Cancel events □ Change staffing levels □ Watch local news for weather information □ Call your local National Weather Service/Monitor NWSChat □ Deploy trained weather spotters □ No impact
13. How are your operations impacted when the area you serve is placed under a Particularly Dangerous Situation (PDS) Watch? * □ Watch the weather forecast more closely/Monitor Radar □ Send information out □ Begin briefings □ Cancel events □ Change staffing levels □ Watch local news for weather information □ Call your local National Weather Service/Monitor NWSChat □ Deploy trained weather spotters □ No impact
REFERENCES League, C. E., W. Diaz, B. Philips, E. J. Bass, K. Kloesel, E. Gruntfest, and A. Gessner, 2010: Emergency manager Albright, C. S., and W. Winston, 2015: Business Analytics, decision-making and tornado warning communication, Data Analysis and Decision Making, 5th Edition. Meteorol. Appl., 17 (2), 163–172. CrossRef. Cengage Learning. 990 pp. Mann, H. B., and D. R. Whitney, 1947: On a test of whether Baumgart, L. A., E. J. Bass, B. Phillips, and K. Kloesel, 2008: one of two random variables is stochastically larger than Emergency management decision making during severe the other. Ann. Math. Statist., 18 (1): 50–60. CrossRef. weather. Wea. Forecasting, 23, 1268–1279. CrossRef. Morss, R. E., and F. M. Ralph, 2007: Use of information Corfidi, S. F., 1999: The birth and early years of the Storm by National Weather Service forecasters and emergency Prediction Center. Wea. Forecasting, 14, 507–525 managers during CALJET and PACJET-2001. Wea. CrossRef. Forecasting, 22, 539–555. CrossRef. Davis, J. M., A. R. Dean, and J. L. Guyer, 2010: A detailed Palisade Inc., 2020: Stat Tools. [Available online at analysis of SPC ‘‘high risk’’ outlooks, 2003–2009. www.palisade.com/stattools/default.asp]. Preprints, 25th Conf. Severe Local Storms, Denver, CO, Storm Prediction Center, cited 2020: About the SPC. Amer. Meteor. Soc. 8B.6. [Available online at: [Available online at www.spc.noaa.gov/misc/aboutus. ams.confex.com/ams/pdfpapers/176214.pdf]. html]. Ernst, S., D. LaDue, and A. Gerard, 2018: Understanding ____, 2021: Frequently Asked Questions. [Available online emergency manager forecast use in severe weather at: www.spc.noaa.gov/faq]. events. J. Operational Meteor., 6 (9), 95–105. CrossRef. Weaver, J., L. C. Harkabus, J. Braun, S. Miller, R. Cox, J. Grice, G. K., and Coauthors, 1999: The golden anniversary Griffith, and R. J. Mazur, 2014: An overview ofa of the first tornado forecasts. Bull. Amer. Meteor. Soc., demographic study of United States emergency 80, 1341–1348. CrossRef. managers. Bull. Amer. Meteor. Soc., 95, 199–203. Hoss, F., and P. Fischbeck, 2018: Use of observational CrossRef. weather data and forecasts in emergency management: Weiss, N., 2008: Introductory Statistics, 8th Edition. Pearson. an application of the theory of planned behavior. Wea. 881 pp. Climate Soc., 10, 275–290. CrossRef. Jirak, I. L., C. J. Melick, A. R. Dean, S. J. Weiss, and J. Correia Jr., 2012: Investigation of an Automated Temporal Disaggregation Technique for Convective Outlooks during the 2012 Hazardous Weather Testbed Spring Forecasting Experiment. Preprints 26th Conf. Severe Local Storms, Nashville, TN, 10.2. [Available online at: ams.confex.com/ams/26SLS/webprogram/ Paper211733.html].
ISSN 2325-6184, Vol. 9, No. 3 46